[libvirt] [PATCH v2] Get the maxvcpus from both the qemuCaps and /dev/kvm

Jiri Denemark jdenemar at redhat.com
Thu May 5 06:55:49 UTC 2016


On Wed, May 04, 2016 at 10:44:37 -0400, Cole Robinson wrote:
> On 05/04/2016 10:24 AM, Jiri Denemark wrote:
> > Hmm, I think it is a very bad idea to limit the CPUs according to
> > something which is hidden and you cannot influence. In other words, the
> > API itself is not able to report the maximum depending on emulator
> > binary, machine type, whatever so it shouldn't do that at all. Using
> > default emulator and machine type is even more confusing that reporting
> > just the kernel limits.
> 
> Is it not confusing that MaxVCPUs will return a value that no VM can actually
> support, because it hits qemu limits? That was the original complaint.
> 
> Maybe instead we do MIN(kvm_vcpus, max_vcpus_supported_by_any_qemu_machine)
> 
> Even if we don't change the logic, the API documentation could use some work
> to clarify the existing behavior

I think this is the only part that really needs changing. There's no way
we can fix the API to report anything useful. The result will always be
confusing in some way. I'd rather keep the confusion consistent and let
the API report what it always reported and explain that in the
documentation with a pointer to domain capabilities API.

Jirka




More information about the libvir-list mailing list